The jiffies tests in n_tty_receive_overrun() are simplified ratelimiting
(without locking). We could use struct ratelimit_state and the helpers,
but to me, it occurs to be too complex for this use case.
But the code currently tests both if the time passed (the first
time_after()) and if jiffies wrapped around (the second time_after()).
time_is_before_jiffies() takes care of both, provided overrun_time is
initialized at the allocation time.
So switch to time_is_before_jiffies(), the same what ratelimiting does.
Signed-off-by: "Jiri Slaby (SUSE)" <jirislaby@kernel.org>
Link: https://lore.kernel.org/r/20230827074147.2287-5-jirislaby@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
struct n_tty_data *ldata = tty->disc_data;
ldata->num_overrun++;
- if (time_after(jiffies, ldata->overrun_time + HZ) ||
- time_after(ldata->overrun_time, jiffies)) {
+ if (time_is_before_jiffies(ldata->overrun_time + HZ)) {
tty_warn(tty, "%d input overrun(s)\n", ldata->num_overrun);
ldata->overrun_time = jiffies;
ldata->num_overrun = 0;